The aim of this effort is to illuminate the relationship between multiple forms of investment in children and children's development and wellbeing. Our research agenda is defined by an ecological model of child development, which posits that investments should be defined broadly, including not only family income, but such additional inputs as parental time, emotional warmth and nurturance, monitoring of activities, and consumer goods and services (such as child care and health care). In addition, community level inputs, such as the quality of schools and neighborhood safety, are expected to affect the development and wellbeing of children, as are public policies such as the generosity of income supports and child support enforcement policies. More effective strategies for investing in children can be developed when we understand how these multiple factors work together to shape development of children over the course of childhood and adolescence. Two research agendas are proposed: (1) a set of concrete research tasks investigating the ways in which family-,community-, and policy-level factors combine to sustain children and promote their healthy development, and (2) a broader research plan developed in concert with an interdisciplinary network focused on systematic explorations of topics related to the wellbeing of children and families. For our individual research plan, we propose to conduct multivariate analyses using two longitudinal data bases a) three waves of the National Survey of Children, a data set rich with family process measures, augmented with contextual data at the state and zip code level, and b) data from the Child and Family Subgroup Study, a longitudinal evaluation of the Federal JOBS program among 2,500 mothers with preschool children who receive AFDC. Analyses will address shortcomings of existing research including a tendency to focus on problem behavior rather than positive adaptation, particularly in adolescence; a failure to look across sets of related problem and pro-social behaviors; a tendency to extract developmental patterns from an accumulation of cross-sectional studies rather than longitudinal research; a need for more work on nationally representative-samples; a lack of research on racial and cultural sub- groups; and a lack of research that provides clear guidance to policy makers and program providers. For the work of the collaborative research network, we suggest four goals: l) basic research on individual, family, community, and policy factors that contribute to positive child development; 2) evaluation research on programs and policies that can enhance family and child wellbeing; 3) methodological research to compare and improve the measures available to assess family and neighborhood processes and child outcomes at varied ages and in varied cultural groups; and 4) trend analysis to track a broad range of indicators of child and family wellbeing over time, across population groups, and across nations. Moore and her colleagues at Child Trends look forward to working with other investigators to build a body of research that can inform both public policy and future research.